ESTRO 2025 - Abstract Book

S2187

Interdisciplinary – Education in radiation oncology

ESTRO 2025

Whitney U- and Fischer’s exact tests were used to compare scores and frequency of key treatment terms, respectively.

Results:

All LLM chatbot responses were in the same language as the question, despite no explicit instruction during prompting, and included at least 1 treatment recommendation. The Table shows the scored results overall and per diagnosis. Clinical trials were recommended in 12/104 English and 1/104 Spanish responses (p=<0.01). Specific radiotherapy techniques such as 3DCRT, IMRT, SBRT, and proton therapy were mentioned in 8/104English and 3/104 Spanish responses (p=0.21). Palliative/supportive care was mentioned in 23/104 English and6/104 Spanish responses (p<0.01). Conclusion: The same cancer treatment question, when asked in different languages, could elicit meaningfully different responses from LLM chatbots. In this study, asking a question in Spanish led to responses that were less likely to be guideline-concordant or introduce clinical trials/palliative approaches to treatment as compared to English. Thus, there is a need for robust multilingual evaluation to harness the promise of LLMs in oncology to avoid perpetuation of misinformation and better promote health literacy to diverse patient audiences.

Keywords: artificial intelligence, chatbot, education

References: Chen S, Kann BH, Foote MB, et al. Use of Artificial Intelligence Chatbots for Cancer Treatment Information. JAMA Oncol. 2023;9(10):1459–1462. doi:10.1001/jamaoncol.2023.2954

2951

Digital Poster Breaking down the barriers: challenges and strategies for gender equity in radiotherapy and oncology Raffaella De Pietro Department of Radiation Oncology, Policlinico Umberto I, Sapienza University of Rome, Rome, Italy